24results about How to "Accurate explanation" patented technology

The invention relates to the technical field of synchronous positioning and mapping, in particular to a three-dimensional mapping method and system. Extracting feature points of the obtained target map image by using an ORB-SLAM system, and performing three-dimensional visual reconstruction based on the feature points to generate an initial sparse point cloud map; fusing the sparse point cloud data of the target map image acquired by the plurality of sensors to obtain dense point cloud data, and processing the obtained dense point cloud data by using a point cloud segmentation algorithm and a target detection algorithm to obtain target information in the dense point cloud data; and performing spatial segmentation and organization on the dense point cloud data according to the target information in the dense point cloud data by using a tree data structure algorithm to obtain a point cloud map with a hierarchical structure. According to the method, the map data is accurate, and the management of the map data can be efficiently realized.

The invention discloses an engine group fault mode identification method and device based on a density clustering-support vector machine and multi-moment prediction weighting and electronic equipment, and the method comprises the steps: carrying out the dimension reduction processing of training engine performance parameters and testing engine performance parameters based on principal component analysis (PCA); training engine low-dimensional performance parameters and testing engine low-dimensional performance parameters are obtained; performing fault mode clustering processing on the low-dimensional performance parameters of the training engines based on a density clustering DBSCAN algorithm in predicting the single-moment fault mode of the engine to obtain fault mode classification labels corresponding to the training engines; establishing a single-moment fault mode classification model based on a support vector machine (SVM), and training the SVM-based single-moment fault mode classification model by using the low-dimensional performance parameters of the training engines and the fault mode classification labels corresponding to the training engines to obtain a trained SVM-based single-moment fault mode classification model; inputting low-dimensional performance parameters of the test engines into the trained SVM-based single-moment fault mode classification model to obtain a single-moment fault mode classification label of each test engine; and obtaining a final fault mode classification label of each test engine based on multi-moment fault mode weighted prediction and the single-moment fault mode classification label of each test engine.

The invention discloses a fault diagnosis method based on time sequence characteristic linear modulation intelligent oversampling, and the method supplements the details of a high frequency band through the improvement of the signal sampling rate, and improves the fault diagnosis performance. The method comprises the steps of obtaining an original vibration signal; an original vibration signal is preprocessed, and a low-sampling-rate signal is obtained through down-sampling. And pairing the low-sampling-rate signal and the original vibration signal for intelligent super-sampling network training. And a super-sampling rate model is constructed, and prediction output of super-sampling rate signals is realized. And constructing a fault diagnosis model, and learning features of signal fault classification. And inputting the training set into a super-sampling rate fault diagnosis model to train the model. And inputting a vibration signal into the trained model to obtain an oversampling rate signal, and classifying the oversampling rate signal to obtain a health state type of the oversampling rate signal. The fault diagnosis model provided by the invention better solves the problem that the diagnosis performance is poor when facing low-sampling-rate signals, and meanwhile, the high-frequency information of the signals is relatively truly recovered.

The pull-type transient electromagnetic detection system comprises a detection module, a detection assembly comprises a transmitting coil, a receiving coil and a host, the transmitting coil and the receiving coil are arranged in parallel, the height between the transmitting coil and the receiving coil is adjustable, one end of the host is connected with the transmitting coil, and the other end of the host is connected with the receiving coil. The horizontal relative position of the receiving coil and the transmitting coil can also be adjusted, the position of the receiving coil is determined when the mutual inductance of the transmitting coil and the receiving coil is zero, the signal processing module comprises an amplifier and a noise processing module, and the receiving coil, the amplifier, the noise processing module and the host are sequentially connected. Signal processing is carried out through the following steps: 1, coil laying; 2, setting an amplifier circuit; 3, electromagnetic noise suppression; according to the invention, in the electromagnetic signal acquisition process and the electromagnetic signal processing process, the quality of the electromagnetic signal is improved, so that the electromagnetic signal with higher quality can be comprehensively obtained.

The embodiment of the invention provides a medical visual question answering method and device based on causal reasoning and a large language model, and is applied to the field related to computer vision, and the method comprises the following steps: S1, respectively extracting initial visual features and initial text features from a medical image and a question text; s2, performing feature mapping operation on the initial visual features; s3, performing feature fusion on the local visual features obtained in the step S2 and the initial text features; s4, performing feature fusion on the fusion feature, the global visual feature and the local visual feature to obtain a visual intermediary variable; carrying out feature fusion on the fusion feature and the initial text feature to obtain a text intermediary variable; s5, performing causal intervention operation; s6, generating corresponding question and answer pairs according to the training samples; and S7, generating a medical question and answer result. In this way, the technical problem that a current medical visual question answering method has some limitations such as lack of relevance or accuracy of question answering is solved.

The invention discloses a fracture-vug type gas reservoir well testing model with a shaft connected with a large karst cave and an establishing method and application of the fracture-vug type gas reservoir well testing model. The method comprises the following steps: selecting proper gas reservoir parameters to establish a fracture-vug type gas reservoir well testing model for connecting a shaft with a large karst cave according to a physical model for connecting the shaft with the large karst cave; according to the gas reservoir characteristics, an energy equation between the shaft and the large karst cave is established, the energy equation considers pressure loss at the joint of the large karst cave and the shaft, and the actual geological condition is better met. The novel fracture-vuggy gas reservoir well test interpretation method using the fracture-vuggy gas reservoir well test model comprises the following steps: performing solution and numerical inversion according to the fracture-vuggy gas reservoir well test model to obtain a real space bottom hole pressure solution; a real space bottom hole pressure value is obtained by utilizing the real space bottom hole pressure solution and combining the gas reservoir parameter actual value; and fitting the real space bottom hole pressure value with the actually measured bottom hole pressure data to obtain parameters such as the volume of the large karst cave. The karst cave volume of the gas reservoir can be explained more accurately by using the gas reservoir type well testing model.

The invention provides a mammary gland ultrasonic focus interpretable diagnosis system, a storage medium and electronic equipment, and relates to the technical field of medical image artificial intelligence auxiliary diagnosis. According to the method, image attribute feature tags are mapped to corresponding nodes of a knowledge graph. Then, the graph neural network model carries out multi-hop reasoning on a graph structure through a message passing mechanism; according to the method, not only can information from related attribute nodes be aggregated, but also the reasoning result of the graph neural network model can be calibrated and explained by utilizing clinical expert rules extracted from the knowledge graph, so that a clear and traceable reasoning path is attached to each prediction of the graph neural network model. The technical problem that image information is separated from the knowledge reasoning process in an existing knowledge graph fusion ultrasonic focus diagnosis method is solved, the visual perception ability of the image and the symbol reasoning ability of the knowledge graph are combined, and therefore an intelligent diagnosis system which is accurate in prediction and clear in explanation is provided.

The invention relates to the technical field of video data analysis and processing, and discloses a video data-based artificial information checking, collecting and early warning method, which comprises the following steps of: S1, collecting video data from cameras mounted at different positions through an identification system; s2, preprocessing the collected video; s3, key features of the preprocessed video are extracted; a predetermined risk area and different types of predetermined transaction areas are divided based on a scene view, risk areas are dynamically divided according to historical data and a real-time monitoring result, key monitoring is performed on a high-risk area, the pertinence and effectiveness of risk management are improved, manual information check and collection are performed on behavior tracks with scores higher than a predetermined threshold, and the risk management efficiency is improved. And the result is fed back to the system in time, and subsequent analysis and early warning are adjusted and optimized, so that the intelligent level and accuracy of the early warning system are improved, the overall safety and reliability of the monitoring system are improved, and potential safety risks are effectively prevented and processed.

The invention relates to an interpretable image segmentation method based on a graph matching strategy, and the method comprises the steps: selecting an image segmentation backbone network, and training the image segmentation backbone network; visual words with semantic information are extracted from the input image through the image segmentation backbone network; according to the visual words with the semantic information, establishing a feature unit-level semantic component relation graph and an overall category-level semantic component relation graph one by one spatial position; and measuring the similarity between the feature unit-level semantic component relation graph and the category-level semantic component relation graph and predicting a segmentation result of the image. The method has the beneficial effects that the reasoning of the deep neural network is converted into an interpretable deductive reasoning process through an interactive matching process between the instance level graph and the category level graph, so that the transparency of model decision is enhanced, and the credibility of a user to model prediction is improved in the key application field.

The application discloses a kind of causal common sense knowledge base construction methods based on graph attention mechanism, comprising the following steps: 1) obtain causal common sense knowledge triples from open knowledge base as training resources;2) using knowledge graph embedding technology, obtain the vector representation of causal common sense knowledge triples;3) combined with background knowledge graph, construct causal common sense knowledge base construction model based on graph attention mechanism, select appropriate loss function to optimize model parameters;4) using causal common sense knowledge base construction model to score the confidence of missing triples, the highest score triples are added to the knowledge base as new learning causal common sense knowledge.The application is based on background knowledge graph, and constructs causal common sense knowledge base construction model using graph attention mechanism, intends to learn new causal common sense knowledge through a small amount of training resources, expands causal common sense knowledge base, so that the knowledge base can better support intelligent question and answer system.

The invention belongs to the crossing field of artificial intelligence and geophysics, and particularly discloses a transient electromagnetic forward modeling method and system considering an induced polarization effect based on deep learning, and the method comprises the steps: inputting Cole-Cole key parameters representing the induced polarization effect into a CNN-LSTM mixed deep learning network model for forward modeling, and obtaining transient electromagnetic response data; wherein Cole-Cole key parameters representing the induced polarization effect serve as multi-dimensional input features of the CNN-LSTM mixed deep learning network model, transient electromagnetic response data of a corresponding time channel serve as output, and the CNN-LSTM mixed deep learning network model is trained by adopting a segmented weighted mean square error loss function. According to the method, the segmented weighted mean square error loss function is adopted in the training stage, higher weights are given to the symbol inversion points and the prediction errors of the adjacent time windows, and the reliability of all-time response prediction is guaranteed.

The invention provides a spectrum inversion high-resolution processing method based on strong axis stripping, and relates to the field of seismic exploration, and the technical scheme is that the spectrum inversion high-resolution processing method based on strong axis stripping is characterized by comprising the following steps: inputting original seismic data, and extracting strong reflection wave event seismic data; subtracting the extracted strong reflection wave event seismic data from the original seismic data to obtain seismic data after strong reflection suppression; replacing the original seismic data with the seismic data subjected to strong reflection suppression, and establishing a spectrum inversion objective function; and solving the spectral inversion objective function to obtain a high-resolution seismic section which suppresses strong reflected waves and highlights weak signals. The method has the advantages that a spectral inversion high-resolution processing mechanism based on strong axis stripping is clear, and time domain iteration deconvolution stripping of the strong reflection wave event, spectral inversion high-resolution processing stripping of the strong reflection wave event, strong reflection suppression and spectral inversion high-resolution thin layer recognition are involved.

The embodiment of the present invention provides a total organic carbon logging interpretation method and system based on the heterogeneity characteristics of shale, including: obtaining relevant data of shale layer sections; obtaining the distribution of homogeneous unit types with depth in each single well coring section according to the core description results and core sample test data in the modeling data; for each homogeneous unit type, using a machine learning algorithm to establish a homogeneous unit logging identification model; using a machine learning algorithm to establish a respective TOC logging interpretation model for each homogeneous unit type; using the drilling data set to be predicted, using the homogeneous unit logging identification model to identify the homogeneous unit types of different depth sections of a single well, and conducting TOC logging interpretation of shale layer sections according to the TOC logging interpretation model of each homogeneous unit. By constructing a rock phase type in which other rock components are relatively homogeneous except TOC, TOC logging interpretation models are established for different rock types, so that TOC interpretation is more accurate.

The invention discloses an online geological survey method for a live-action three-dimensional digital outcrop, and relates to the technical field of crossing of network geographic information and live-action three-dimensional visualization. The method comprises a geological outcrop data co-construction and sharing service logic process, an online geological investigation service logic process and a user geological outcrop data management service logic process, and the co-construction and sharing service logic process comprises a user registration and login sub-process, a geological outcrop data co-construction sub-process and a geological outcrop data sharing sub-process. The investigation business logic process comprises a geological outcrop model and related data uploading sub-process, a geological outcrop model searching sub-process, a geological outcrop model interaction sub-process, a geological outcrop related data display sub-process and a geological information interaction sub-process; and the management service logic process comprises a contact person management sub-process, a geological outcrop data collection management sub-process and a geological outcrop data contribution management sub-process, so that the effectiveness of online investigation of the live-action three-dimensional digital outcrop can be achieved.

A system for providing a purchase incentive has a detection device for detecting a behavior of a person surrounding the detection device, the system being designed to automatically trigger a signaling action related to providing the purchase incentive based on the detected behavior of the person.

The invention discloses a blood pressure measuring method, wearable equipment and a storage medium, and relates to the technical field of wearable equipment, and the blood pressure measuring method comprises the following steps: acquiring a target physiological signal detected by a target user in a current time period based on the wearable equipment; collecting detection environment parameters of the wearable device in the current time period; according to the target physiological signal, blood pressure change factor information which causes the change of the blood pressure value level of the target user is identified, and a target blood pressure calculation model mapped by the blood pressure change factor information is determined from a plurality of preset blood pressure calculation models; and inputting the acquired blood pressure baseline characteristics and blood pressure baseline states, the target physiological signals and the detection environment parameters into a target blood pressure calculation model, and outputting a blood pressure value of the target user. Interference caused by individual differences is effectively reduced, and accuracy and individual adaptability of blood pressure monitoring are improved.

This invention discloses a method for interpreting non-uniform fracture parameters based on fracture propagation simulation forward modeling and pump shutdown pressure drop analysis inversion. The method includes: (1) obtaining geological parameters and construction parameters of the target well, and obtaining pressure data based on the fracturing construction process; (2) using cepstral analysis to identify the opening status of multiple perforation clusters in a single fracturing section based on high-frequency pressure data, and determining the location of the effectively opened perforation clusters among all perforation clusters; (3) using low-frequency pressure data to obtain the total area of ​​the first fracture in a single fracturing section based on the inversion theoretical model; (4) establishing a fracture propagation simulation model, using the location of the effectively opened perforation clusters and the total area of ​​the first fracture as physical constraints, and simulating the total area of ​​the second fracture. When the difference between the total area of ​​the second fracture and the total area of ​​the first fracture is less than a set threshold, the interpretation of non-uniform fracture parameters in a single fracturing section is achieved. This invention can achieve rapid and accurate interpretation of non-uniform fracture parameters.

This invention discloses a method for intelligent greeting, explanation, and lead generation in architectural exhibition halls using a digital human based on a Finite State Machine (FSM), belonging to the field of 3D virtual interaction and intelligent customer acquisition technology. The invention utilizes a real-time user behavior acquisition unit to collect 3D roaming behavior in milliseconds. The FSM control unit incorporates five mutually exclusive states and a standardized transition matrix, driving the digital human to intelligently switch between idle, greeting, precise explanation, lead generation guidance, and order placement states. The digital human driving unit employs lightweight rendering and lip-synced broadcasting for large architectural scenes. A scene-product linkage unit achieves native linkage between explanation and transaction. An intelligent guidance unit gently acquires high-intent leads. A data iteration unit optimizes interactive effects in a closed loop. This invention solves the industry pain points of traditional 3D exhibition halls, such as unattended operation, traffic loss, stiff interaction, disconnect between explanation and transaction, and inability to automatically acquire customers and generate leads. It achieves 365-day unattended intelligent operation, significantly improving user dwell time and business conversion rates. It is applicable to all scenarios, including 3D cloud exhibitions in the architectural industry chain, digital exhibition halls for building materials companies, industrial park investment promotion, and government and enterprise displays for urban renewal.

The invention provides a method and device for quickly constructing a horizontal well formation interpretation model. The method comprises: determining a reference well according to horizontal well data; selecting a formation curve for square wave processing to determine a depth value and a square wave back curve value; performing environmental correction on the square wave back curve value using an environmental correction plate to determine the best approximate value of the formation as an initial attribute value, and establishing an initial formation model in combination with the depth value; obtaining formation interface data, and intersecting with the well trajectory profile of the initial formation model as a reference line drawn into the well trajectory profile; forming a layer interface using multi-well gridding differences, and intersecting with the well trajectory profile, and determining a constrained initial formation model in combination with adjacent well projections; forward modeling the constrained initial formation model according to the response of a logging instrument to determine a forward modeling curve and a forward modeling formation model; comparing the forward modeling curve with the measured curve, and repeating the forward modeling when there is inconsistency between the two curves until they are consistent, and outputting the horizontal well formation interpretation model.

The invention provides an artificial intelligence interpretation method for multi-arm hole diameter logging data. The artificial intelligence interpretation method for the multi-arm hole diameter logging data comprises the steps that 1, imaging processing is conducted on logging interpretation data; 2, building a working condition interpretation model based on the convolutional neural network; step 3, training the built working condition interpretation model; 4, the well trained working condition interpretation model is utilized, and according to the collected actual multi-arm hole diameter logging data, the multi-arm hole diameter logging working condition is judged. According to the artificial intelligence interpretation method for the multi-arm hole diameter logging data, by combining the high-precision data acquisition capability of multi-arm hole diameter logging and the efficient intelligent analysis advantage of artificial intelligence, the interpretation efficiency and interpretation precision of multi-arm hole diameter logging interpretation work are effectively improved, and the labor intensity of workers is reduced.

The invention discloses a rare word theme mining system based on LLM, and relates to the technical field of text analysis and information extraction, and the technical scheme is characterized in that the rare word theme mining system comprises a term extraction module, a semantic space positioning module and a theme distribution module; the invention aims to combine the advantages of a seed-guided topic discovery method and LLM to more accurately identify and mine specific topics related to rare words, the system overcomes the limitation of processing the rare words in the prior art, the ability to understand the meanings of the rare words is improved through an innovative method, the efficient topic discovery process is realized, and the method is suitable for large-scale popularization and application. New breakthrough and innovative application are brought to the fields of text analysis and information extraction, and users are assisted to better understand and mine hidden values in text data.

This application provides a method and system for bandgap prediction and interpretation based on a multi-head self-attention mechanism. The method includes: using a Transformer-based encoder to model the global dependencies of a CSV file containing the physicochemical characteristics of lithium-ion battery materials through a multi-head self-attention mechanism, determining the context embedding, and concatenating it with the physicochemical characteristics of the lithium-ion battery materials to determine training features; then using a pre-set CatBoost regressor to perform regression training on the training features to determine the Catformer model; using the Catformer model to predict the bandgap of the lithium-ion battery material to be tested, and using the SHAP analysis method to perform interpretability analysis on the bandgap prediction. This application enhances the global dependency characteristics of the material, achieves accurate bandgap prediction, and reveals the physicochemical significance of the prediction results, providing a general tool for the rapid design of lithium battery electrode materials.

The present application provides a financial platform risk prediction method and system based on multi-source user behavior data, which obtains multi-source user behavior data indicated by the financial platform risk prediction task, and uses graph autoencoding technology to process the data to generate graph autoencoding structure data, thereby performing a preliminary assessment of user behavior in the preliminary risk prediction stage to generate first risk prediction data, and then determining the influence weights of different features on risk prediction based on the influence feature domain and probability score of the graph autoencoding structure data, and then using the influence weights output by the target graph autoencoder to converge the encoded data to generate feature interpretation data, and finally generating an attention weight distribution based on the feature interpretation data, and blending it with the original multi-source user behavior data to generate blended data, and performing advanced risk prediction based on the blended data to generate more specific and accurate second risk prediction data. In this way, user behavior risks can be comprehensively and accurately assessed.

The invention provides a text simplification method and system based on retrieval enhancement generation, and relates to the technical field of text simplification. According to the method, the to-be-simplified sentence is obtained, and the complex words in the to-be-simplified sentence are recognized; retrieving a plurality of possible paraphrases of the complex words, respectively replacing the complex words in the to-be-simplified sentence with the possible paraphrases to form a plurality of paraphrasing sentences, and selecting the paraphrasing sentence with the highest similarity with the to-be-simplified sentence from the plurality of paraphrasing sentences as an optimal paraphrasing sentence; filling the sentence to be simplified and the optimal paraphrasing sentence into a predefined prompt template to obtain an input sentence converted by the prompt template; and taking the input sentence as input data of the large language model, and outputting the simplified sentence. According to the method, the complex text paraphrases related to the semantics of the content to be simplified are retrieved, so that the text simplification accuracy can be improved, and the relevance of the text and the integrity of information can be ensured to be kept.